
Uncertainty-Aware Reach-Avoid Estimation with Statistical Guarantees
Predicts reach-avoid outcomes or abstains, with finite-sample bounds on the error rates of its predictions.
I work on data-driven reachability analysis for verifying robot controllers. Instead of relying on a model of the controller, I use its own executions to estimate the states from which it reaches its goal without failing, and how likely it is to do so when the dynamics are noisy. These estimates can be calibrated to carry statistical guarantees. Because they only need data, they work for black-box controllers such as reinforcement learning or behavior cloning policies, whose internals can't be analyzed directly.
This builds on my earlier work in the lab on planning and control for robots with complex dynamics, and on estimating controllers' regions of attraction from data.
PhD student in Prof. Kostas Bekris's PRACSYS Lab, Rutgers University




Predicts reach-avoid outcomes or abstains, with finite-sample bounds on the error rates of its predictions.

Estimates reach-avoid probabilities under stochastic dynamics from a flow-matching terminal-state model.

A follow-up to the 2024 survey that reviews methods combining explicit models with implicit, learned ones. It sorts them into four families by how tightly the two are coupled: sequential pipelines, parallel residuals, hierarchical decompositions, and embedded architectures.
My roleCo-wrote the taxonomy of integrative methods, finding the work that belongs under each family and proposing new splits within them.

A trajectory-tracking controller for robots with arbitrary dynamics: it keeps the robot on its planned trajectory despite noisy observations by jointly estimating its recent trajectory and adapting its controls. Unlike pure tracking, it can deviate from the plan to avoid obstacles. Estimation and control run online at 10 Hz or more.

A survey of 50 years of methods for generating robot motion, from those built on explicit models to those that learn implicit ones, and the opportunities to combine them.
My roleCo-wrote the sections on explicit motion planning and control, reinforcement learning, and behavior cloning.

Executes planned trajectories safely with only a rough dynamics model tuned from data, combining kinodynamic replanning, feedback control, and a safety mechanism, for mobile robots on surfaces with unknown friction.
My roleCo-developed the method, proposed its contingency-checking safety mechanism, and ran the physical experiments on the MuSHR platform.

Speeds up planning for robots with complex dynamics by connecting a learned controller through a precomputed roadmap. Tested on vehicles driving over uneven terrain and on a quadrotor.

Estimates from data where a robot controller succeeds, known as its region of attraction, by analyzing its dynamics in a learned latent space. It scales to a 67-dimensional humanoid and a 96-dimensional three-fingered manipulator.
My roleProposed and implemented a label-based contrastive training variant.

Classifies land cover in multifrequency radar (PolSAR) images using autoencoder features and superpixels.




NSF-NRT SOCRATES Fellow (2024 – 2026)

Thesis: Safety and Efficiency for High-dimensional Robots: Integrating Learning, Planning, and Control
Outstanding Project and Outstanding Publication Awards from Rutgers CS for MORALS (2024)

Expected to graduate in May 2029, and open to internships and collaborations.